The making of: The Two Towers (a 25 driver Full Range line array)

Hi Kolby,

Pleased to meet you and I'm happy to discuss all things electrical, I find it fun and invigorating! Send me a PM any time. Since wesayso has moved on and I don't wish to muddy up the thread further I'll give a few last comments and leave it :).

We are not even close to discover ALL the parameters that the ear/brain uses...

Agreed, it is always good to strive to investigate further. That said we have come quite a ways in the last 100 years with our knowledge of how the ear works and have developed a number of metrics for measuring things such as DACs and Amplifiers. What has not changed that much, if at all is how our ear works. At least we do not have fast moving goal post!

...and that is why, IMHO , things can measure exactly the same , but sound different

We will have to agree to disagree at this point which is ok:). With todays metrics and measurement gear it is very very hard to come by with 2 pieces of equipment (DAC and/or AMP) that have different designers that measure exactly the same. I would go so far to say it is very hard with the same model, same PCB, same circuit and same manufacturing run to get 2 units to measure exactly the same with todays measurement gear and known metrics.

I also believe that we are able to measure things that are below the threshold of audibility. Gear (DAC/Amps) that measures very very close in a large number of areas that are level matched and double blind tested I believe would be very hard to pick out. There is a reasonable amount of material written by Toole, Geddes and Harman (which they have graciously shared) that support this view but I concede there are still unknowns that could sway the results; however, that would be the rare case and not the norm (maybe an item that could contribute to the community?).

Since this is DIY audio we thrive on the rare case and that is ok as that is where we usually learn something new! There is always room to learn something new:). Measuring and quantifying the entire reproduction chain including the room is an insurmountable accomplishment, I am reduce to the electrical side of things.
 
Its been a long time since I did anything with op amps. For small signals, they are near perfect amplifiers - infinite gain, no phase shift, etc . so the feedback around the op amp determines the transfer function. For large signals and high frequencies, you can bump into their limitations - like slew rate limit. A long cable could make that worse if it looks like a capacitive load to the driver. So for understanding what is happening it might help to look at the specs of the op amp chip driving the cable, if available.

For the M1 DAC:

Like many of Musical Fidelity's earlier solid-state designs, the M1DAC's output stage is based on 5534-type op-amps, in this case one JRC 5532 dual op-amp chip per channel, used as a differential line driver. This decades-old op-amp has always offered low static distortion and the ability to drive low impedances, as can be seen in fig.8. The M1DAC was driving the punishing 600 ohm load for this spectral analysis, but the highest-level distortion harmonic (the subjectively innocuous second) lies at –96dBFS in the left channel (blue trace, 0.0015%) and –89dBFS in the right (red, 0.003%). The next-highest harmonic (the third) lies at –114dBFS in the left channel (0.0002%) and –99dBFS in the right (0.001%). This is superb output-stage linearity, and increasing the load impedance to 100k ohms didn't change the picture by much. Intermodulation distortion was similarly low (fig.9).
 
I wouldn't call 600 ohms punishing. 100 ohms would be and that is the ballpark for twisted pair. A shield would lower the Z. Not saying this is root cause but we are looking for 2nd order effects. If you could,do a the sweep into 100 ohms and saw increased thd, that would be a clue. So far,just speculating
 
On the Xonar Essence ST they used the following opamp:
As you can see by the above THD Vs frequency graph the LM4562NA keeps excellent THD ratings across the audible range. Depending on the design and circuit load the LM4562 can produce THD ratings as low as 0.00003%. This chip is also very quick with a slew rate of +/- 20V/µs (typ) and a GBP of 55Mhz. These opamps are used as buffers on both ST and H6 cards for the line outputs.

Isn't that the same/similar as used in the Universal Buffer?
Universal Buffer – Neurochrome
 
JRC 5532 not the last word in quality op-amps IMHO

Probably not but from looking at the data sheet, its really quite adequate. OTOH, I'd love to compare its data sheet with what you consider a really good one. I know if I were designing for myself I would pay for a premium device; if I were designing for mass production perhaps not.

Data sheet referenced it as being a high performance version of 1458 op amp. Back in the 70's, 1458 was my bread and butter dual op amp for commodity audio apps. 40-50 years later should find much better devices!

I just have one picture from the data sheet that illuminates previous discussion
attachment.php


Upper left graph shows no significant phase shift until well past 20 khz

Upper right graph shows output limitations with loads under 300 ohms or so. How does this matter? Well cable looks like 100 ohms to the wavefront, which traverses it in a few microseconds, after which the driver sees the input impedance of the amp at the end of the cable. I don't think that is significant for audio. perhaps there could be distortion generated if the cable got long enough.

Lower left graph shows maximum output voltage swing vs frequency. It can swing rail to rail up to 100 khz. Perfectly adequate for audio. Only issue I would see is if the rails weren't far enough apart - i.e. the power supply voltages weren't high enough for the output swing needed.

I don't know enough to say anything about the noise graph but just below I see a teaser title of a graph I'm not showing - THD vs output swing but its a few hundredths of a percent for an audio signal.

Sounds pretty good to me. If you have 5532's in your equipment, I wouldn't worry about them.
 

Attachments

  • 5532 op amp.jpg
    5532 op amp.jpg
    213.2 KB · Views: 298
Thought about you when I read this article....

For me, big chills towards the end of Vangelis - Friends of Mr. Cairo.... every time.

If Listening To Music Gives You Chills You May Have A Unique Brain | IFLScience

I definitely fall into that category, and always assumed it was potentially a similar experience for everybody?
I don't even need a perfect source to experience it, it is the musical content that triggers it but it does help to have a good source available.

Isn't this quite common among the listeners here on our forum? Honest question :).
 
Probably not but from looking at the data sheet, its really quite adequate. OTOH, I'd love to compare its data sheet with what you consider a really good one. I know if I were designing for myself I would pay for a premium device; if I were designing for mass production perhaps not.

Data sheet referenced it as being a high performance version of 1458 op amp. Back in the 70's, 1458 was my bread and butter dual op amp for commodity audio apps. 40-50 years later should find much better devices!

I just have one picture from the data sheet that illuminates previous discussion

Upper left graph shows no significant phase shift until well past 20 khz

Upper right graph shows output limitations with loads under 300 ohms or so. How does this matter? Well cable looks like 100 ohms to the wavefront, which traverses it in a few microseconds, after which the driver sees the input impedance of the amp at the end of the cable. I don't think that is significant for audio. perhaps there could be distortion generated if the cable got long enough.

Lower left graph shows maximum output voltage swing vs frequency. It can swing rail to rail up to 100 khz. Perfectly adequate for audio. Only issue I would see is if the rails weren't far enough apart - i.e. the power supply voltages weren't high enough for the output swing needed.

I don't know enough to say anything about the noise graph but just below I see a teaser title of a graph I'm not showing - THD vs output swing but its a few hundredths of a percent for an audio signal.

Sounds pretty good to me. If you have 5532's in your equipment, I wouldn't worry about them.

Lets first state that the M1 DAC does not sound bad, just running it like I have done for the Goldmund I probably eat up too much internal processing space to have it shine. In comparison the Xonar DAC does a better job, though it is a different job. The M1 is an upscaling DAC which could explain some of the difference heard.
However, for both DAC's I experience better results with the JDS Atom added as a pré-amp. It sort of brings back the dynamics that are there at a reduced volume level.
 
I definitely fall into that category, and always assumed it was potentially a similar experience for everybody?
I don't even need a perfect source to experience it, it is the musical content that triggers it but it does help to have a good source available.

Isn't this quite common among the listeners here on our forum? Honest question :).
not me, I'm more of a savage beast whose soul gets soothed
 
I definitely fall into that category, and always assumed it was potentially a similar experience for everybody?
I don't even need a perfect source to experience it, it is the musical content that triggers it but it does help to have a good source available.

Isn't this quite common among the listeners here on our forum? Honest question :).

I have to rectify this a little bit. Thinking about it a bit more, chills down the spine most often have happened when I'm being immersed in audio. Like in a good venue with a great band playing, but I also (quite often, actually) have experienced this at home and even in my Car, which has been prior focus of mine for getting good/great sound.
Now goose bumps, that can happen with more mediocre equipment. The chills may have become more commonplace than I realized at home :rolleyes:.

There is a difference, obviously. Although some on the forum might still be of the opinion I'm listening to TV speakers here :D.
 
I have this "unique" brain as well. I think it might explain why some people are obsessed with hifi. When I first heard about this not being a universal thing everyone experienced I was quite surprised. I find this happens more frequently when I have good sound. For me it is usually music I like with time signature changes or dynamics.